SOME TYPICAL ECONOPHYSICS’

AND SOCIOPHYSICS’ MODELS

Gheorghe SĂVOIU*

Abstract. Some typical Econophysics’ and Sociophysics’ models result from

this new sciences’ way of thinking or from the physicist’s methods used in
other domains as Economics and Sociology. This paper describes the
improvements of the quality in the classical research of Economics and
Sociology through some of these new and still typical models and tries also
to investigate why econophysicists’ and sociophysicists’ models are able to
perform financial or sociological analysis, and which are their most
interesting strengths and weaknesses. Econophysicists’ and Sociophysicists’
models seek to integrate the Physics’ methods and laws with classical
Economics’ and Sociology’s theory and thinking, seeing this new domain of
applied Physics as an unlimited one. Econophysics and Sociophysics replace
conventional ways, with the new and broader views of Physics’ thinking. The
author believes that an important scope and intention of this paper is to
draw a repertory of some typical models for the use of Economics and
Sociology. In addition to this main purpose, the paper could be a statistical
evaluation of some not so typical phenomena as crises or recessions, which
can be reordered along the new coordinates of contemporary Physics’
thinking and specific models.
Keywords: Statistical Physics, Econophysics’ model, Sociophysics‘ model,
power law, diffusion, weak and strong signals.

1. Introduction

Econophysics and Sociophysics describe applications of Physics to

different fields, similar to Astrophysics, Geophysics, and Biophysics. The
specific fields or domains are in these distinctive cases Economics and
Sociophysics.
Thus, Econophysics is an “interdisciplinary research field applying
methods of statistical Physics to problems in Economics and Finance” [1].
The contemporary way to define Econophysics is to do so in terms of the
*
University of Piteşti, 1 Târgul din Vale, St., Piteşti, Romania, e-mail: gsavo-
iu@yahoo.com, gheorghe.savoiu@upit.ro

55
ideas that it involves in effect physicists doing Economics with theories
from Physics, this raises the question of how the two disciplines relate to
each other and it explains interest rates and fluctuations of stock market
prices, these theories draw analogies to earthquakes, turbulence, sand piles,
fractals, radioactivity, energy states in nuclei, and the composition of
elementary particles.
Sociophysics is a new insight into the applicability of much of
elementary statistical physics to the social sciences. Sociophysics means a
new insight followed by transferring and further developing ideas and
concepts common to Physics, Biology, and Ecological Systems. First
named Psychophysics, Sociophysics can be described as the sum of
activities of searching for fundamental laws and principles that characterize
human behaviour and result in collective social phenomena. Sociophysics
tries to model the dynamics of social and economic indicators of a society
and investigate how life extension will influence fertility rates, population
growth and the distribution of wealth [2], religion, friendship and sex,
social network, traffic etc. Sociophysics has become an attractive field of
research over the last two decades, despite the controversies between
sociophysicists and sociologists. Its relevant potential used for understan-
ding the social phenomena always will win.
Econophysics and Sociophysics improve the quality of the classical
research of Economics and Sociology through their original models. New
models, already called typical after only five years, result from a new way
of thinking or from the trans-disciplinary methods used in new domains.

2. The scientific research model

The expansion of contemporary science has multiplied their number
to over 1,000 independent sciences, especially within borderline areas (e.g.
econophysics, situated at the border between physics and economics,
sociophysics – at the border between physics and sociology etc.). Science
emerges when at least three elements are joined together: a distinctive
theory, a segment of reality as a specific object, and a model interposed
between theoretical investigation and its object of study. Sciences have
their own characteristic models and laws, acquired mainly thanks to their
inclination for measuring their object of study.
From the tetragrams of the ancient Chinese culture to abstract or
geometrical figures in ancient Greece, from the first music sol-fa systems
to the everyday languages used by calculus programs, all these types of

56
presentation laying special emphasis on the logical element of a visual
nature have been, and are still, simplified alternatives to modelling.
In a relevant way, the model and modelling have been situated,
through their initial practical uses, close to geometry, than any other
scientific domain. The appearance of the term as such is linked to the year
1868, when the mathematician Eugenio Beltrami managed to construe an
early Euclidian model for non-Euclidian geometry.
For the first time, he was turning the model and modelling into a
concept, studying, by their agency, “a domain, a phenomenon, an object
inaccessible to direct research”. The geometry-inspired model became “a
coagulant factor” for scientific thinking, a continuous process of ponde-
ring, represented, symbolized and conveyed, no less than the tetragrams
were to Gottfried von Leibniz the inductive solution to the mechanic
system of his own calculating mechanical device. At a higher level of
elaboration, models are scientific representations, or representations of
scientific theories. Paraphrasing Parmenides, the model that can be
thought, and the one for which the thought exists are one and the same.
Theoretical science, a permanent source of experimental suggestions,
becomes at once experimenting and foreseeing, and along these lines the
basic conditions of multi-dimensional modelling can be synthesized as
follows:
– the first condition for a model is its direct relationship with thinking
(“a bird is a machine functioning in accordance with the laws of
mathematics, an instrument that man can reproduce with all its motions” –
to quote Leonardo da Vinci, in Macchine per volare);
– a second condition is the identification of the essential aspects, and
formulating questions in a correct manner;
– the profoundness, the intensity, and the depth represent the third
condition of representation through models (the oscillation between
analogy and he convention-symbol);
– the efficiency of the transposition, or the translation of the
theory into the reality of the world under study seems to be anther
condition, the superior models becoming themselves objects of research
and re-modelling [3].
In keeping with the reasoning of modelling, as maximum fidelity
translation or transposition, any theory corresponds to a model, and any
model, when validated through the agency of reality, will correspond to
reality. However, the closer the model will draw to the point of intersection
of several sciences, the more correct the transposition/translation. Even the

57
exclusive answer to the question “what is a model” constitutes a difficult
undertaking, and needs many-sided approaches. Below are some
illustrative variants:
– in the option of physics, a model is a calculating instrument, with
the help of which one can determine the answer to any question concerning
the physical behaviour of the system in question, or else a precise pattern
of a certain segment of the physical reality (two examples, which are today
as well-known as to become banal, are the modelling of the inertial
reference system, and the atomic model);
– in the vision specific to chemistry, the model becomes a structural
concept that attempts to explain the properties found experimentally, or a
support in deductively passing from the general to the specific, a knowing
instrument that forecasts facts and “indicates the numbers” (as in the
memorable example of Mendeleyev’ s table of elements, or the periodicity
of chemical elements);
– in the approach of biology (genetics), a model is considered a
natural modality – reproduced experimentally – of genetically differen-
tiating the populations (the model of DNA being, in this respect, a
commonly cited example, and a relevant point in case);
– in the perspective of mathematics, the model is superposed to a
certain type of measuring methods, specific to mathematical research, with
a view to explain, in an objective manner, the “manner n which the micro-
components and their mutual interactions, either interpreted individually,
or grouped in subsystems, generate and explain the whole of the system”
(Octav Onicescu and the model of informational energy), or a “definition
and non-contradictory description of a number of processes and
phenomena”, of the theses, postulates and axioms, as well as their logical
and mathematical correspondence;
– from a logical point of view, within the structure of the model, the
causes equalize the effect (Anton Dumitriu);
– from a behaviourist standpoint, the model presupposes a number of
participants gathered in a formal way, who “maximize their utility by
starting from a stable set of preferences and accumulate an optimal amount
of information in a variety of markets” (Becker’s model);
– along the lines of the semantic, linguistic and explanatory
dominant, the model is a theoretical or material system by means of which
one can study, indirectly, the properties and transformations of a different,
more complex system, where the first system exhibits an analogy
(according to the explanatory dictionary);

58
 – in its statistical acceptation, the phase-directed sense of the concept
of model is that of a link in an integrated process of knowing, and is made
up of a hypothesis, a schematic representation of a process (phenomenon),
the statistical testing, and the resuming of the process in a general theory;
– in keeping with modern sciences, the multidisciplinary model
becomes the optimum instrument for solving a number of complex general
problems, and modelling turns into a series of means meant to disclose the
real nature of the problems, where the isolated vision does not allow one to
formulate characteristic laws.
– the statistical & mathematical or statistical & physical type of
modelling is a mathematical transcription of a number of simplified
hypotheses about the state or evolution of a social-economic phenomenon,
or physical system under the factorial influence of variables that are
physical or can be assimilated to the physical ones (in the modern scientific
vocabulary, a statistical model also designates the explanatory hypothesis –
model: χ2, F, t etc.).
The Econophysics’ or Sociophysics’ models turn to account the
language and methods of mathematics, testing and statistical decision, the
pattern of physics in assessing (quantum, thermodynamic, acoustic etc.)
reality, as well as the real variables of the segment subject to research
(money flow in the economy, human behaviour in sociology etc.).
How can one manage to practically construct a model? The starting
point is direct experience, or unmediated contact with reality. In order that
a theory could be turned an experiment, or into an “organized contact with
reality”, a theory is formulated, which is subsequently represented by a
material, intuitive or symbolic model, as a filtered reflection of reality.
Louis Pasteur would elegantly underline the primacy of the theory, through
the agency of the well-known formula: “luck favours only the well-
prepared minds”. Tiberiu Schatteles used to synthesize the likeness
between theory and modelling through the phrase “the dogmatics of
isolation”. For instance, economic models are always partial models, and it
is hence always possible to add another equation, quite irrespective of how
large the system is. [4] In order to illustrate a phenomenon, the theory
isolates it from the contingent, very much as the experiment is underlain by
a type of material (i.e. laboratory) isolation. Studying a phenomenon in
isolation also presupposes defining the framework of the isolation through
postulates or axioms as “something that goes without saying”. Due to their
isolation tendencies, Econophysics and Sociophysics are somehow
sciences of models joined to the art of choosing models which are relevant
to the contemporary world. It is compelled to be this because, unlike the

59
typical natural science, the material to which it is applied is, in too many
respects, not homogeneous through time. The object of a model is to
segregate semi-permanent or relatively constant factors from those which
are transitory or fluctuating so as to develop a logical way of thinking
about the latter, and of understanding the time sequences to which they
give rise in particular cases… In Physics and in other natural sciences the
object of experiment is to fill in the actual values of the various quantities
and factors appearing in an equation or formula; and the work when done
is once and for all. In classical economics, theoreticians think that this is
not the case, and to convert a model into a quantitative formula is to
destroy its usefulness as an instrument of thought… “To do so would make
it useless as a model. As soon as the model is done, it loses its generality
and its value as a mode of thought” [5, 6].
Modelling, as a complex iterative process, oscillates between
simplified variants like the “triad” (formulating a hypothesis, collecting the
experimental material, and verifying the hypothesis), and excessively
detailed variants (formulation of the initial model followed by the forming
of repartition classes, gathering the experimental material or the data,
choosing a particular repartition, checking the degree of concordance of the
repartition chosen with the real situation and formulating the hypotheses
that explain the random mechanisms that have generated the data). The
typological diversity of the models results from the great number of the
scientific theories that they reproduce. Seen from the angle of the aim they
were created for, the models fall in two major types: the category of the
rational or theoretical models, and the category of the operational models,
or prediction (decision-making) models.
Through comparison with the time variable, modelling is static or
dynamic. A major classification of modelling according to the typology of
the explanatory variables reveals the deterministic type of modelling in the
past (evolution of phenomena, determined solely by the mechanical, or
simply causal variables) and modern probabilistic modelling (which
contains perturbing variables, in keeping with the probable effect of some
uncontrolled factors and unspecified variables). Contemporary evolutions
of Econophysics and Sociophysics led us to the idea that modelling can not
be only multidisciplinary. For a succinct description of modelling in
Econophysics and Sociophysics, a few clarifications are in order, relating
to their architecture and paradoxes, but also to their various stages.
The architecture of multidisciplinary modelling capitalizes on:
a) minimal simplification through hypotheses (it was formulated for
the first time by William Ockham as the first economic architecture, or of

60
parsimony) or the existence of a minimal number of propositions not
connected mutually, and undemonstrated propositions (out of two interpre-
tations of a phenomenon, the interpretation having fewer supposetions or
simplifying hypotheses is preferred);
b) he simple alternative (the highly intricate models failed to lead to
categorically better results, as against the simple extrapolation formulas –
Koopmans T.C.);
c) the value certified through the dialectical reasoning (a model
facilitates the discussion, clarifies he results and limits the reasoning
errors);
d) the cultural component (if the humans’ economic and social
actions were independent of their cultural inclinations, the enormous
variability of the economic and social configuration in point of time and
place could by no means be accounted for);
e) the shifting from only one discipline to many sciences or to a
multidisciplinary model, through successive models (improvement through
imitation, through analogy, and through passing from one type to another
in Econophysics and Sociophysics).
Modelling in Econophysics and Sociophysics remains a process with
a paradoxical content. The paradox of the infinity of the multivariable
system is revealed by the infinite number of factors, which cannot be
classified in a direct manner, in proportion to the particular model
construed out of a finite number of essential factors. The paradox of the
“relative reduction of one system to the next” proceeds from relative
reducibility, centred on the translatability of the languages concerning
various fields of reality, and manifests itself as an antithesis between the
functional and the substantial. The paradox of the “unique community” can
be translated through the antinomy holding between the correlation of the
action of several models, and the building up of a unique model for a given
problem. The paradox of the “double idealisation” concerns the phases of
simulation, and respectively, of the assignation and interpretation of
information within the model. Multiplied, information is not lost from the
model, very much as, fragmented, it is nothing but information. The
“double idealisation” consists in treating information as “signification” of
information in the model of attribution, whereas in the interpretation model
it is treated equally as signification and as sense.
The concrete stages of modern modelling in Econophysics and
Sociophysics are the following:
1. the structural defining of the system (isolating the phenomenon,
formulating the questions, identifying the major interest variables),

61
 2. the preliminary formulation (sets of hypotheses and conclusions
concerning the relationships between the variables), collecting the empi-
rical (relevant) data,
3. the estimation of the parameters and of the functional forms,
4. the preliminary (gross) testing,
5. the additional testing (based on the new data),
6. the decision – accepting or rejecting (in conditions of predictions
conforming or failing to conform with the available empirical evidence).
Synthetically, the relationship between completeness and preci-
sion/accuracy generates specific models (Table 1):

Table 1
Degree of the data’s completeness and precision generating
the typology of the model

Degree of completeness Degree of precision Typology

of the data of the data of the model
Maximum Maximum deterministic
Relatively low relatively high probabilistic
Relatively high relatively low fuzzy
Relatively low relatively low intuitive
Minimum Minimum nondeterministic

The algorithm of the model has three characteristic features: determi-

nism in point of performance, succession in point of operation, universality
in so far as the spatial, temporal and structural entries and limitations are
concerned. Modelling exhibits three main ways of analysis:
– using the equilibrium equations between the factors, from
Leontief’s input-output balance, to the fuzzy ones, and to those of quantum
physics, thermodynamics etc. (Leontief’s model was subsequently genera-
lized in three distinct variants, i.e. the deterministic, the random, and the
information ones, the fuzzy model has impredictable variation parameters,
the quantum physics’ model means transgresion or transition from energy
to light, or from wave to particle etc.);
– identification of the extreme values as the model of the
“catastrophes”, or of R. Thom’s “critical points” – which is the frequently
cited example in point;
– construction or simulation of conflict situations through the
“strategic games with incomplete information (i.e. competitive situations)
or complete information (i.e. open situations).

62
 The uncertainty of decision-making is paramount, all the way from
Wald’s (prudent or pessimistic) model, characterized by choosing the
maximum profit variant, or the minimal loss cost wise, in the most
unfavourable situation, to Laplace, which selects the higher average-profit
variant, or the lower average-loss, in the hypothesis that the states have the
same occurrence probability, to Savage, where an option is made for the
lowest possible regret (i.e. the usefulness lost as a result of selecting a
different variant than the optimal one, in conditions of complete
information), and to Hurwicz, whose coefficient of optimism re-enters,
through its real-value interval, the vast realm of the probabilities, namely
[from 0 to 1].
To illustrate the above, multidisciplinary modelling maximizes the
capacity of reducing the degree of imprecision/inaccuracy and of assessing
that imprecision/inaccuracy through statistical testing and testing in terms
of probability theory, whereas even mathematical modelling approximates,
while failing to express reality exactly as it is, because reality is not
“exact/precise”, but subject to the stochastic laws or to the action of the
law of great numbers.
To express in a Econophysics and Sociophysics manner how
inaccurate a model is, is more important than modeling in a classical,
hence sophisticated, isolated manner, lacking the power of specificity. The
perspectives of the field of model-construction astonish through the rigour
of a new concept, namely that of the system of Econophysics and Socio-
physics as multidisciplinary models, which presupposes the following
principles:
• the human decision has the fundamental role in its functioning;
• the construction is a logical succession, and also a process of
arrangement in time, in keeping with the principle of economy, or
the law of parsimony;
• the separation and combination of the individual models occurs in
procedure-based chains;
• the system stays open, thus facilitating the adding / the deletion of
restrictions and variables;
• the physical-mathematical structure is independent of the manner
of utilization;
• the architecture is modular, hierarhical and dynamic;
• the information-based and logical connections are, in turn, part of
cooperative, hierarchical, mixed models;
• although including different types of models, the database if
unique.

63
 In the natural harmony of the Econophysics and Sociophysics’
approach to modelling [7], the contribution scored by discovering
of an original model is to be considered much higher than
knowing a new phenomenon or process.
The limits of classical sciences’ modelling are abvious:
• no classical model can consistently and substantially incorporate
the residual variables and areas (which can occasionally be quite
considerable in point of proportions and significations);
• both human behaviour and other random variables like the climate,
the radical political evolutions such as the revolutions, etc., as soon
as they are modelled, bestow an increased amount of uncertainty to
the respective model;
• the model has evolved in a credible manner along the coordinates
of the chronological series, and less so, however, along those of the
territorial series, of the associated / correlated series, in the specific
situations of value optimizations, or concerning verisimilar,
attainable targets developing programmes.

To conclude, a model can be said to represent an image of a specially

selected part of reality, with the aid of which answers can be given to
various questions, or problems belonging to an assortment of fields in the
area of scientific knowledge can be solved, with a certain degree of realism
and a certain limit of error. The main disadvantage of the classical model,
if one resorts to the example provided by the very econometric one, is
revealed by the lack of accuracy of their prediction, by the representatives
of the neoclassical Austrian school of economics Ludwig von Mises and
Friedrich von Hayek. The sad balance of the predictions made by the
econometric models over the past few years, for all the modern calculation
equipment added to the sophisticated classical or uni-disciplinary models,
is nothing but an additional confirmation [8, 9].
Econophysics and Sociophysics’ models are nothing else than partial
models, and it is hence always possible to add another equation, quite
irrespective of how large the system is. What seems to be required, then, is
a determination of the reliability of the estimates of a given system within
the system, but without adding further equations. Classical model are
limited and modern through their multi-disciplinary solutions are more
adequate to reality.

64
 4. Some typical models
from Econophysics and Sociophysics
Econophysics means also a scientific approach to quantitative
economy using ideas, models, conceptual and computational methods of
statistical physics. In recent years many of physical theories like theory of
turbulence, scaling, random matrix theory or renormalization group were
successfully applied to economy giving a boost to modern computational
techniques of data analysis, risk management, artificial markets,
macroeconomics [10].
In Econophysics, the activities of research focused on economic
phenomena but are analyzed by concept, method and model of physics.
Here three typical examples are:
a) the derivation of a price’s distribution in the stock market (the
change in the price “x” of stock market could be considered a random
among dealers, then can derive a diffusion equation as a Brownian motion,
for distribution f(x, t) of price in the stock market) [11]:
∂f ( x, t ) 1 ∂ 2 f ( x, t )
= × .
∂t k ∂x 2
b) distributions of the form that follows a power law as:
ln p ( x) = −α ln x + C , where the constant α is called exponent of the power
law, and C is constant and mostly uninteresting (once α is fixed, it is
determined by the requirement of normalisation to 1), or in the case of
taking the exponential of both sides, this is equivalent to: p( x) = Cx − α
(a power-law distribution occurs in an extraordinarily diverse range
of phenomena such as Finance, Macroeconomics, Demography’s urba-
nism) [12].
c) a fractal and chaos analysis originating as Benoite Mandelbrot
pointed out that the change in the price of the stock market has a fractal
structure for certain range of time interval [13,14], and characterized as a
self-similar structure expressed as: x(t) = CtD, where D is a fractal
dimension, calculated by the box counting method. The fractal structure is
special case of a chaos and chaotic behaviour is very common in a non-
linear system as for an economic system; whether the process is chaotic or
not can be determined by sign of Lyapunov index λ defined as: λ = 1/ n Σ
log | F ′(t ) |, and when λ is positive (negative) then the process is chaotic
(non-chaotic) [10].
Modern Econophysics has developed a new learning system for
econophysicists, a system consisting of several methodological parts:

65
 1) Basic Mathematics’ methods,
2) Basic Econometrics’ methods,
3) Echonophysics’ methods, including chaos’ methods and fractals’
methods,
4) Virtual market’s methods.
reviewing classical methods and concepts concerning to each part:
Mathematical representation and analysis of the economic data for basic
Econometrics; the chaos and fractal including the Lyapunov index and the
fractal dimension for Econophysics; the Sato-Takayasu model and
simulation for virtual market [10].
A very difficult problem in the specific modelling is the testing of the
data, model or of the predictions based on modeling. The question is
always the same: How well can past information predict future evolutions?
The main assumption is that there should be no pattern in the time series of
information, or with other words, the information should be approximated
by a random walk (and the autocorrelation of the information time series
should be negligible). Sometimes the theories of modeling are called or
„baptized” with very strange names as they can be found in the literature
(i.e. in the efficient market hypothesis there are three major models named
the fair-game model, the martingale or sub martingale model and the
random walk model [15].
Sociophysics aims at a Statistical Physics modelling of large scale
social phenomena, like culture and opinion formation and dynamics,
cultural and behavioural dissemination, the origin and evolution of
language, competition and conflicts, crowd behaviour, social contagion,
gossip and rumours evolutions, Internet and World Wide Web, cooperation
and scientific research, appearances of terrorism etc. A good overview of
several fields of application and an accessible, entry-level description of
many simulation models can be interpreted as forming part of the
Sociophysics. For instance, in a paroxysm crisis of fear, opinions can be
activated very quickly among millions of mobilized citizens, ready to act in
the same direction, against the same enemy, but a lot of phenomena can be
studied within the new emerging field of Sociophysics, in particular the
dynamics of minority opinion spreading, the rumour propagation, etc
[16,17,18]. The most remarkable pioneers of Sociophysics probably are
Serge Galam (Sociophysics: a personal testimony), Dietrich Stauffer
(Sociophysics Simulations I: Language Competition), Paris Arnopoulos
(Sociophysics: Chaos and Cosmos in Nature and Culture). The list is
necessarily limited and unavoidably lacking of many important contri-
butions in this research area [19].

66
 In the last two or three decades new interdisciplinary approaches to
social science have been developed by natural scientists. The distribution
of unemployment required a new understanding of society, the dynamics
of social systems has been gradually introduced by W. Weidlich (1972)
and H. E. Stanley (1992) and a thermodynamic approach to social pro-
blems has been favoured by D. K. Foley (1994), J. Mimkes (1995),
A. Drăgulescu and V. M. Yakovenko (2001).
For a better understanding, there are detailed some models of
spreading opinions within a human population. Serge Galam was the first
who have modelled the spread of opinions within a population and gets an
equation of the inertia of democratic systems against changes. In the last
twenty years, sociophysicists have introduced a series of Sociophysics
models. These could be divided in different general classes, which deal
respectively with:
a) opinion dynamics,
b) decision making,
c) competitions / conflicts, fragmentation versus coalitions,
d) income or wealth spreading and concentration,
e) residential segregation, migration dynamics,
f) cultures and languages evolution,
g) friendship and sex,
h) internet and world wide web evolution,
i) religion spreading,
j) social networks dynamics,
k) traffic dynamics,
l) democratic voting in bottom up hierarchical systems,
m) terrorism spreading etc.
Using these original models several major real political social and
religious events were successfully predicted (from the victory of the
French extreme right party in the 2000 to the voting at fifty-fifty in
Germany or Italy). The models are real important tools for a reasonable
perspective and make Sociophysics a predictive solid field. Sometimes
model are philosophical instruments more than scientific. In the year 2000,
Katarzyna Sznajd-Weron have proposed another model of opinion
formation, which was based on trade union maxima “United we Stand,
Divided we Fall” (USDF) known as the model (SM). The main
characteristic of SM model is that information flows only outward. A great
hope for the model of Sociophysics is to show similar correspondence
between simple interactions among entities (agents being the preferred
sociophysical term) and complex behaviour in the final aggregate. A

67
generalized model of opinion formation in a sociophysical way details in a
mathematical way the spread of thinking through social groups. In the
hypothesis of a system consisting of N individuals (members of a social
group), in which each of them can share one of two opposite opinions on a
certain subject, denoted as: σi = ± 1, i = 1,2,…,N.
The opinions of the individuals may change simultaneously
(synchronous dynamics as in Glauber theory) in discrete time steps
according to the rule:
 exp(−li / T )

 δi (t ) with the probability exp(−li / T ) + exp(li / T )
 
δi (t + 1) =   .
− δi (t ) with the probability  exp(li / T )
 
 exp(−li / T ) + exp(li / T )
In this model parameter „T” could be called the “social temperature”
and I means the impact that determines the individual person to change his
opinion when Ii > 0 [20,21]. Since the 1780s, when Euler invented network
theory, till nowadays many models and applications of the graph theory
[22] exist under the form of network analysis. A model, proposed by
Ausloos, M., Gligor, M., [23] considers that the M agents (countries)
which the ME time series refer to, may be the vertices of a weighted
network. The weight of the connection between i and j reflects the strength
of correlations between the two agents and can be simply expressed as:
wij (T ) = | Cij (T ) |, fulfilling the obvious relations: 0 ≤ wij ≤ 1; wij = wji and
wij = 1 for i = j. One must stress at this point that the link connecting the
vertices i and j does not reflect here either an underlying interaction.
Instead, the weight wij is a measure of the similarity degree between the
ME fluctuations in the two countries. The term “fluctuations” refers here to
the account of the annual rates of growth of the considered ME indicator.
Networks are characterized by various parameters. For instance, the vertex
degree is the total number of vertex connections. It may be generalised in a
M
weighted network as: K i = ∑ X ij (j being different from i). Thus, the
j =1

1 M M
average degree in the network is: 〈 K 〉 = ∑ ∑ Wij (j being different
M i =1 j =1
from i). Finally, neither the model itself could solve everything, nor the
method could analyse all the details. But there is a major conclusion in
Sociophysics, that in modelling the human group’s behaviour, a crucial

68
point always remains to study the group decision making and the related
issue of the collective opinion formation and dynamics.

4. A final remark and conclusions

The real criticism of Econophysics is the absence of age variable,
because models of Econophysics consider immortal agents who live
forever, like atoms, in spite of evolution of income and wealth as functions
of age, that are studied in economics using the so-called overlapping-
generations models. Sociophysics needs more clarity, especially when it
envisions probability at the foundation of social theory. There is no
contradiction between this new field of Sociophysics and the Statistics.
But, certainly, sociophysicists should be more careful when they are
justifying their complex models. Sometimes this minds action seems to be
averaged out and finally removed by virtue of the law of large numbers.
To conclude, Econophysics and Sociophysics models are multi-
disciplinarity models and they try to unify, while classical models remains
uni-disciplinarity models and they have succeed only to isolate. Thence,
the culture of multidisciplinary modelling remains a practical issue, not
certainly in as far as that culture is regarded only as a product of life, but
life (reality) having become, in that sense, a consequence or an imprint of
culture ... The model needs the three great intellectual faculties,
perception, imagination, and reason, and most of all he needs imagination,
to put him on the track of those events which are remote or lie below the
surface, and of those effects of visible causes which are remote or lie below
the surface. (Alfred Marshall)
The typical models of Econophysics and Sociophysics were and still
remain the results of the weak or of the strong signals coming from
outside, from the reality, into science’s thinking.

REFERENCES

[1] Yakovenko, V. M. (2007), Econophysics, Statistical Mechanics Approach to, Encyclo-

pedia of Complexity and System Science, Springer http://refworks.springer
[2] Mantegna, R. N. Stanley, H. E. (2000), An Introduction to Econophysics: Correlations
and Complexity in Finance, Cambridge University Press, Cambridge, pp. VIII-IX.
[3] The researchers who confine themselves to a sole model burden temselves with an
even greateer sin, as Georgescu-Roegen remarked: “the sin of complete ignorance of
the qualitative factors that cause endogeneous variability”.
[4] Schlicht Ekkehart (1985), Isolation and Aggregation in Economics, Springer-Verlag,
Berlin, Heilderberg, New York, Tokyo, p. 60

69
[5] Keynes, J. M. (1936), The General Theory of Employment, Interest and Money.
Macmillan, London, pp. 44 and 297.
[6] Keynes, J. M. (1973), The General Theory and After, Part U: Defense and
Development. The Collected Works, vol. XIV, Macmillan, London, (3) pp. 285-332
and pp. 296-299.
[7] Pecican, E. Ş. (2003), Econometric for... Economists (in Romanian), The Economical
Publishing House, Bucharest, pp. 46-47.
[8] Ludwig von Mises (1966), Human Action: A Treatise on Economics, 3rd rev. ed.
Chicago, Henry Regnery and Co.
[9] Hayek, Friedrich (1989), The Collected Works of F.A. Hayek. University of Chicago
Press.
[10] Burda, Z., Jurkiewicz, J., Nowak, M. A. (2003), Is Econophysics a solid science?, Acta
Physica Polonica B3487.http://www.arxiv.org/abs/ cond -matpapernum/0301096,
[11] Cui, L., Yamada, K., Kaburagi, Kang, M. (2005), WEB Based Learning System for Eco-
nophysics, July 7-9, 2005, Juan Dolio, Dominican Republic, IEEE, 0-7803-9141-1/05,
pp. 1-2.
[12] Mandelbrot, B. B. (1963), The Variation of Certain Speculative Prices. Journal of
Business, 36, pp. 394-419.
[13] Mandelbrot, B. B. (1983), The Fractal Geometry of Nature, W. H. Freeman and Co.
[14] Mandelbrot, B. B. (1997), Fractals and Scaling in Finance, Springer-Verlag.
[15] Săvoiu, Gheorghe Andronache, Constantin (2009), On The Efficiency Of Financial
Markets The 33 rd ARA Congress, Modernism and Progress in Arts and Sciences, June
02-07, 2009 Sibiu, http://www.uamsibiu.ro
[16] Galam, S. (2008), La Science Magique et le Rechauffement du Climat, Ed. Plon, Paris
Press.
[17] Galam, S. (2002), Minority Opinion Spreading in Random Geometry, European
Physical Journal B 25, pp. 403-406.
[18] Galam, S. (2003), Modeling Rumors: The No Plane Pentagon French Hoax Case,
Physica A 320, pp. 571-580.
[19] Galam, S., Jacobs, F. (2007), The role of inflexible minorities in the breaking of
democratic opinion dynamics, Physica A 381, pp. 366-376.
[20] Kacperski, K., Holyst, J. A. (2000), Phase transitions as a persistent feature of groups
with leaders in models of opinion formation. Physica A 287, pp. 631-643.
[21] Holyst, J. A., Kacperski, K., Schweitzer, F. (2000), Phase transitions in social impact
models of opinion formation. Physica A 285, pp. 199-210.
[22] Biggs, N. L., Lloyd, E. K.,Wilson, R. J. (1976), Graph Theory, pp. 1736-1936. Oxford
University Press, Oxford.
[23] Ausloos, M., Gligor, M. (2008), Cluster Expansion Method for Evolving Networks
having Vector-like Nodes. Acta Physica Polonica A 114(3), pp. 491-499.

70